Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper tackles the challenge of early mental health disorder detection in Spanish by introducing three foundational models tailored to the mental health domain through domain-specific pre-training. It also presents Incremental Context Expansion (ICE), a novel automatic relabeling method that enhances training sample quality by pinpointing when cumulative social media messages indicate a disorder. The combination of these specialized models with ICE significantly improves detection latency and performance on three Spanish benchmarks, marking a notable advancement in the field.
Early detection of mental health disorders in Spanish just got a major boost, cutting detection latency while enhancing accuracy with innovative models and methods.
Early detection of mental health disorders is often limited by the lack of specialized resources in Spanish and the difficulty of analyzing long histories of social media posts. This paper addresses these challenges through three main contributions. First, we introduce three Spanish foundational models specifically adapted to the mental health domain through domain-specific pre-training. Second, we propose Incremental Context Expansion (ICE), an automatic relabeling methodology designed for early detection. ICE identifies the point at which cumulative messages provide enough evidence of a disorder, generating more informative training samples. Third, we provide a set of fine-tuned models using the samples generated with the ICE methodology for early risk detection tasks. Our results on three Spanish benchmarks show that combining these specialized models with ICE improves the state-of-the-art, reducing detection latency while maintaining high performance. All models are publicly available.